def extractDataFromLink(queue, urls, filename, fileCount):
dbObj = mdb.mongodbDatabase('doc_collection')
docs = dbObj.docCollection
down_doc = docs.find_one({'url':urls,'primaryEnt':filename})
if(down_doc == None or (down_doc['documents'] == None) or len(down_doc['documents'])==0):
try:
#print "down load docs for ",urls
cleanText = ''
if(urls.endswith('.pdf')):
print "############# found pdf #############"
proxy_support = urllib2.ProxyHandler({"http":"proxy.iisc.ernet.in:3128"})
opener = urllib2.build_opener(proxy_support)
urllib2.install_opener(opener)
with open('filename','wb') as f:
f.write(urllib2.urlopen(URL).read())
f.close()
content = convert('filename')
cleanText = content.encode('utf-8','ignore')
else:
extractor = Extractor(extractor='ArticleExtractor', url=urls)
extracted_text = extractor.getText()
cleanText = cleanTheExtraction(extracted_text)
sentenceList = tokenizer.tokenize(cleanText) #get sentences
if(len(sentenceList) > minLen): # write to a file if the extraction size is greater than min no. of sentences
curFile = filename+str(fileCount)+'.txt'
senList = []
for l in sentenceList:
newl = l.encode('utf-8','ignore')
senList.append(newl)
document = {'url': urls, 'documents':senList, 'primaryEnt':filename}
if down_doc == None:
post_id = docs.insert_one(document) #.inserted_id
else:
docs.replace_one({'url': urls, 'primaryEnt':filename},document,True)
sentenceString = ' '.join(sentenceList)
getTripleList(sentenceString,urls,filename)# call a function to do corenlp->sentcreate->ollie
except Exception, e:
print "error in boilerpipe code: ",e," url: ", urls
exc_type, exc_obj, exc_tb = sys.exc_info()
fname = os.path.split(exc_tb.tb_frame.f_code.co_filename)[1]
print(exc_type, fname, exc_tb.tb_lineno)
def extractDataFromLink(queue, urls, filename, fileCount):
dbObj = mdb.mongodbDatabase('doc_collection')
docs = dbObj.docCollection
down_doc = docs.find_one({'url':urls,'primaryEnt':filename})
if(down_doc == None or (down_doc['documents'] == None) or len(down_doc['documents'])==0):
try:
extractor = Extractor(extractor='ArticleExtractor', url=urls)
extracted_text = extractor.getText()
cleanText = cleanTheExtraction(extracted_text)
sentenceList = tokenizer.tokenize(cleanText) #get sentences
if(len(sentenceList) > minLen): # write to a file if the extraction size is greater than min no. of sentences
curFile = filename+str(fileCount)+'.txt'
# p = file('/tmp/extractions/'+curFile, 'w')
for s in sentenceList:
try:
if(ord(s) < 48 or ord(s) > 122):
sentenceList.remove(s)
else:
print "@@@@@",s
# p.write(s)
# p.write(" ")
except:
sentenceList.remove(s)
# p.close()
document = {'url': urls, 'documents':sentenceList, 'primaryEnt':filename}
if down_doc == None:
post_id = docs.insert_one(document) #.inserted_id
else:
docs.replace_one({'url': urls, 'primaryEnt':filename},document,True)
sentenceString = ' '.join(sentenceList)
getTripleList(sentenceString,urls,filename)# call a function to do corenlp->sentcreate->ollie
except Exception, e:
# print "whats the error ",e
# print urls
pass
else:
print "@@@@@",s
# p.write(s)
# p.write(" ")
except:
sentenceList.remove(s)
# p.close()
document = {'url': urls, 'documents':sentenceList, 'primaryEnt':filename}
if down_doc == None:
post_id = docs.insert_one(document) #.inserted_id
else:
docs.replace_one({'url': urls, 'primaryEnt':filename},document,True)
sentenceString = ' '.join(sentenceList)
getTripleList(sentenceString,urls,filename)# call a function to do corenlp->sentcreate->ollie
except Exception, e:
# print "whats the error ",e
# print urls
pass
else:
try:
curFile = filename+str(fileCount)+'.txt'
oldVal = docs.find_one({'url':urls,'primaryEnt':filename})
sentenceList = oldVal['documents']
sentenceString = ' '.join(sentenceList)
getTripleList(sentenceString,urls,filename)# call a function to do corenlp->sentcreate->ollie
except:
pass
dbObj.client.close()
